From the Department of Radiology, Columbia University, New York-Presbyterian/Lawrence Hospital, 622 W 168th St, PB-1-301, New York, NY 10032 (K.A.V.); Department of Surgery, Urology Division, City of Hope, Duarte, Calif (S.A.C.); Departments of Radiology (L.S., S.S.R.) and Surgery, Urology Division (S.R.), University of California, Los Angeles, Los Angeles, Calif; and Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, Calif (N.K.).
Radiology. 2018 Dec;289(3):721-727. doi: 10.1148/radiol.2018180786. Epub 2018 Aug 14.
Purpose To evaluate the performance of translabial (TL) US in preoperative detection of sling erosion into pelvic organs with cystourethroscopic and surgical correlation. Materials and Methods The study cohort included women who underwent surgery at a subspecialty center (from 2008 to 2016) for suspected mesh complications in the setting of previous midurethral sling placement for stress urinary incontinence (from 1999 to 2012) with available preoperative TL US imaging. Clinical information, the finding of sling erosion identified intraoperatively and at cystourethroscopy, and blinded dual-reader radiologic analysis of the TL US studies for mesh location (intraluminal, mural, or extramural) relative to pelvic organs (bladder, urethra, vagina, or rectum) were evaluated. The diagnostic performance of TL US was correlated with the reference standard of surgical findings. The consensus of two radiologists was recorded, and interobserver agreement was evaluated with the κ statistic. Results Of the 124 women who were suspected of having sling erosion (mean age, 57.5 years ± 11.1 [standard deviation]), 15 women (12.1%) had sling erosion into the urethra or bladder at surgery. Sensitivity and specificity for erosion at TL US were 53% (95% confidence interval: 45%, 62%) and 100% (95% confidence interval: 97%, 100%), respectively, when erosion was defined as only intraluminal mesh products. Sensitivity and specificity for erosion at TL US were 93% (95% confidence interval: 89%, 98%) and 72% (95% confidence interval: 65%, 80%), respectively, when erosion was defined as visualizing either intraluminal or intramural mesh products. Interobserver agreement (κ value) was 0.95. Cystourethroscopy had 67% sensitivity and 100% specificity for sling erosion. Conclusion Preoperative translabial US can be used to detect sling erosion into the lower urinary tract, with sensitivity up to 93% and specificity up to 100%. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Benson and Phillips in this issue.
目的 评估经阴道超声（TL-US）在经阴道超声术前检测吊带侵蚀入盆腔器官中的性能，与膀胱尿道镜和手术结果进行相关性分析。
材料与方法 研究队列包括在专门中心接受手术的女性（2008 年至 2016 年），这些女性在先前接受过中尿道吊带术治疗压力性尿失禁（1999 年至 2012 年）后出现吊带相关并发症，且有术前 TL-US 成像资料。评估了临床信息、术中发现的吊带侵蚀和膀胱尿道镜检查，以及对 TL-US 研究的双盲放射学分析，以确定吊带位置（管腔内、壁内或壁外）与盆腔器官（膀胱、尿道、阴道或直肠）的关系。TL-US 的诊断性能与手术结果的参考标准相关联。记录了两位放射科医生的共识，并使用κ 统计评估了观察者间的一致性。
结果 在 124 名疑似存在吊带侵蚀的女性中（平均年龄 57.5 岁±11.1[标准差]），15 名女性（12.1%）在手术中发现吊带侵蚀至尿道或膀胱。当将 TL-US 上的侵蚀定义为仅管腔内的吊带产品时，其侵蚀的敏感性和特异性分别为 53%（95%置信区间：45%，62%）和 100%（95%置信区间：97%，100%）。当将 TL-US 上的侵蚀定义为可见管腔内或壁内的吊带产品时，其侵蚀的敏感性和特异性分别为 93%（95%置信区间：89%，98%）和 72%（95%置信区间：65%，80%）。观察者间一致性（κ 值）为 0.95。膀胱尿道镜检查对吊带侵蚀的敏感性为 67%，特异性为 100%。
结论 术前经阴道超声可用于检测下尿路的吊带侵蚀，其敏感性高达 93%，特异性高达 100%。
